aabyaneh2026mit
srlwam
Sim-to-real dataset for Franka Panda two-cube stacking task with simulation data generated in Isaac Lab and real data captured on physical robot cell.
Downloads109
Episodes8200
Why This Matters for Physical AI
Provides calibrated sim-to-real stacking demonstrations with domain randomization and physical verification to enable transfer learning and robustness evaluation for manipulation tasks.
Technical Profile
- Modalities
- rgbdepthproprioception
- Robot Embodiments
- Franka Panda
- Action Space
- joint_positions
- Environment
- simulationlab
- Task Types
- manipulationstackingpick_and_place
- Episodes
- 8200
- Data Format
- HDF5
- Annotation Types
- action_labelsreward_labels
- License
- mit
Access
Need custom rgb data?
Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.
Request a Sample Pack